Discriminant Matrix Research Articles

Ultra-short term wind power prediction method based on LightGBM and transformer

Abstract Accurate ultra-short-term wind power prediction is very important for the safe and stable operation of the power system. At present, most wind power forecasting systems use numerical weather forecast data and the data collected by the Supervisory Control And Data Acquisition (SCADA) system. However, these data have the characteristics of high data dimension, feature redundancy, and feature correlation. Therefore, in the process of ultra-short power prediction, it is necessary to select features with high power correlation and low cross-correlation between features. In this paper, a feature selection method for wind power prediction based on LightGBM is proposed to calculate the importance of all features and determine the preliminary selected features. Then, the maximum information coefficient is used to construct the correlation discriminant matrix, according to which the features with similar importance in a single screening are evaluated, and the input features with high similarity are discarded. In addition, in order to improve the accuracy of the system’s wind power prediction, this paper uses the effective features extracted by LightGBM to design an ultra-short-term power prediction method based on Transformer. The results show that the RMSE of ultra-short-term wind power prediction is 7.3%.

Research on extraction method of new type of train control system operation scenario elements in special environments

At present, the construction of high-speed railway in the western region has become one of the key points in China. However, the existing train control systems do not adapt to the environment. So a new type of train control system (TCS-NT) is proposed. However, for TCS-NT, the definition and content of its operating scenarios are not clear, and the construction of operating scenarios is mostly involves subjective analysis. It faces problems such as a lack of a comprehensive operating environment analysis method, a method for selecting scenario elements for varying test requirements, and inexplicable design choices. In this paper, the Operational Design Domain (ODD) and a key element extraction method for simulation scenarios are proposed to analyze the operational environment of the TCS-NT. Using this method, the impact of each scenario element on the sub-functions is sequentially analyzed in the tracking operation scenario of the TCS-NT. According to the influence on each sub-function, a mapping equation is constructed for the three-dimensional plane comprising element, structure, and function, and scenario elements are extracted using the discriminant matrix. In this paper, the method is used to analyze the operation scenario in the train tracking scene, and the feasibility of the method is verified.

Evaluation of Urban Resource Environmental Carrying Capacity and Land Spatial Development Suitability in a Semiarid Area of the Yellow River Basin

At present, the carrying capacity of resources and the environment in some areas has reached an upper limit, and the problems of ecological destruction and environmental pollution have become increasingly prominent. The overexploitation and disorderly development of land space, resulting in the spatial imbalance between the population economy and the resource environment, is one of the root causes of excessive resource consumption, ecological degradation, and environmental pollution. Resource and environmental carrying capacity and suitability evaluation of land space development is the basis of land space planning. It provides conditions for the modernization of land space governance. The framework provided by the “ Double Evaluation Guide (Trial) “ is universal. It is necessary to adapt it to local conditions and to achieve the ” Double Evalution” results, according to the characteristics of regional development. Taking Hohhot as the research area, which is based on the “Guide”, involves selecting convincing and credible evaluation factors and using the single factor integration method and the discriminant matrix method. The “double evaluation” of Hohhot was also analyzed. This provides basic data for regional high-quality development and land space planning. The results show the following: (1) The ecological function of Hohhot is good. The areas with low importance for ecological protection accounted for about 18%, the areas with medium ecological protection accounted for about 62%, and the areas with high ecological protection accounted for about 21%. (2) The carrying capacity of agricultural function in Hohhot is medium. The carrying capacity of the low polar region accounted for about 23%, the medium polar region’s carrying capacity accounted for about 71%, and the high polar region’s carrying capacity accounted for about 6%. (3) The urban function carrying capacity of Hohhot is medium. The carrying capacity of the low-grade area accounted for about 25%, the medium-grade area’s carrying capacity accounted for about 55%, and the high-grade area’s carrying capacity accounted for about 21%. (4) The overall agricultural production suitability grade of Hohhot is not suitable. (5) The suitability of urban construction in Hohhot is general. The unsuitable area accounts for about 40%, the more suitable area accounts for about 45%, and the most suitable area accounts for about 15%.

Simulation and Zoning Research on the Ecosystem Service in the Beijing–Tianjin–Hebei Region Based on SSP–RCP Scenarios

Understanding future trends and identifying characteristic differences in regional ecosystem services (ESs), in addition to ecological zoning, is vital for promoting the adjustment of ecological policy and the construction of sustainable ecosystems. Based on shared socioeconomic pathways and representative concentration pathways (SSP–RCP), the spatial distribution pattern of land use in the Beijing–Tianjin–Hebei region (BTH) in 2030 was simulated by using a patch-generating land use simulation model (PLUS) in this study. Water yield, carbon storage, habitat quality, and food product were simulated through the comprehensive evaluation model combining ecosystem services and trade-offs (InVEST). The comprehensive supply capacity of ESs was assessed. The ecological risk level was calculated by comparing the changes in the integrated supply capacity of ESs in 2020 and under each SSP–RCP scenario in 2030, and ecological zoning was established using a two-dimensional discriminant matrix. The results are as follows: (1) The degradation of grassland and cropland accompanied by an increase in construction land and forest to varying degrees will be the common characteristics of the three SSP–RCP scenarios in BTH. (2) Water yield and carbon storage services will exhibit an upward tendency only under SSP1-2.6, while habitat quality and food product services will exhibit a downward trend under three SSP–RCP scenarios. Obvious geographical heterogeneity exists in the comprehensive supply capacity of ESs. (3) Zones with low ecological risks will mainly be distributed in some counties of Zhangjiakou city, while zones with high ecological risks will account for a large proportion of the whole. There will be great ecological risks in the BTH overall. (4) The BTH was split into four types of ecological zones based on supply and risk. These zones comprise the ecological enhancement zone, ecological restoration zone, ecological sensitive zone, and ecological conservation zone. Corresponding control measures were also proposed. The findings of this study can be used to inform the formulation and improvement of environmental conservation policies.

Water-sand inrush risk assessment method of sandy dolomite tunnel and its application in the Chenaju tunnel, southwest of China

Sandy dolomite strata are widely distributed in the diversion tunnel of the Yuxi section of the Water Diversion Project in Central Yunnan (WDPCY). The water-sand inrush is the main failure phenomenon in the construction of sandy dolomite tunnel, which has seriously and commonly threatened the safety of the Chenaju (CNJ) tunnel in the WDPCY. In this paper, we established the discriminant matrix of the water-sand inrush risk assessment by consulting the experts, including the designers and construction personnel of the CNJ tunnel. Then, the risk assessment of water-sand inrush in the CNJ tunnel is carried out based on the proposed Fuzzy-AHP method. The results show that (1) a hierarchical model for risk assessment of water-sand inrush in sandy dolomite tunnel is constructed by criterion and index including thirteen factors. (2) The main criterion layer factors affecting water-sand inrush in the sandy dolomite tunnel are dolomite’s sandy degree and water yield. (3) The risk grade of water-sand inrush in the sections of YX49 + 089∼YX49 + 099 and YX51 + 637∼YX51 + 647 in the CNJ tunnel is V, which indicates very high risk. The assessment results are consistent with the actual engineering situation, which indicate the proposed method can provide the reference for the risk management in a similar project. Key Policy Highlights Thirteen main factors induce water-sand inrush in sandy dolomite tunnel. Characteristics of sandy dolomite mass and groundwater condition play a decisive role in sandy dolomite tunnel for water-sand inrush. A water-sand inrush risk assessment system established based on Fuzzy-AHP method is suitable for sandy dolomite tunnel.

Sparse Local Fisher Discriminant Analysis for Gas-Water Two-Phase Flow Status Monitoring With Multisensor Signals

Gas-water two-phase flow has typically stable flow statuses and constantly changing transition flow statuses. Accurate identification and real-time monitoring of flow status are conducive to the in-depth study of two-phase flow and the safe operation of industrial process. A monitoring strategy based on sparse local Fisher discriminant analysis (SLFDA) is proposed in this article. First, multisensor signals are obtained to reflect flow process information. Second, the least absolute shrinkage and selection operator is used to find the sparse discriminant directions to determine the key variables relevant to the flow process from multiple sensor signals. Then, the weight coefficient matrixes of SLFDA keep the original structure of the same flow status data and make the data of different flow statuses more separated, which distinguish different flow statuses to the maximum extent. Finally, two monitoring indexes including the discriminant index and the stability index are established to analyze the dynamic flow process, which enable a concurrent monitoring of both flow evolution and instability to realize fine-scale description of flow process. SLFDA can monitor various flow statuses through only one projection discriminant matrix by transforming high-dimensional signals into features representing the flow characteristics, which avoids model traversal and improves monitoring efficiency. Further study of flow status features provides meaningful physical interpretation and in-depth process analysis with consideration of actual flow process. The application on the data of gas-water two-phase flow in horizontal pipe demonstrates the feasibility and efficacy of the method.

Deep Manifold Harmonic Network With Dual Attention for Brain Disorder Classification.

Numerous studies have shown that accurate analysis of neurological disorders contributes to the early diagnosis of brain disorders and provides a window to diagnose psychiatric disorders due to brain atrophy. The emergence of geometric deep learning approaches provides a new way to characterize geometric variations on brain networks. However, brain network data suffer from high heterogeneity and noise. Consequently, geometric deep learning methods struggle to identify discriminative and clinically meaningful representations from complex brain networks, resulting in poor diagnostic accuracy. Hence, the primary challenge in the diagnosis of brain diseases is to enhance the identification of discriminative features. To this end, this paper presents a dual-attention deep manifold harmonic discrimination (DA-DMHD) method for early diagnosis of neurodegenerative diseases. Here, a low-dimensional manifold projection is first learned to comprehensively exploit the geometric features of the brain network. Further, attention blocks with discrimination are proposed to learn a representation, which facilitates learning of group-dependent discriminant matrices to guide downstream analysis of group-specific references. Our proposed DA-DMHD model is evaluated on two independent datasets, ADNI and ADHD-200. Experimental results demonstrate that the model can tackle the hard-to-capture challenge of heterogeneous brain network topological differences and obtain excellent classifying performance in both accuracy and robustness compared with several existing state-of-the-art methods.

Degenerate scales for thin elastic plates with Dirichlet boundary conditions

A degenerate scale occurs when a loss of uniqueness of the solution of the boundary integral equations happens for that scale of the problem. We consider here the biharmonic 2D problem with Dirichlet boundary conditions which models the bending behavior of a clamped isotropic Kirchhoff plate. We extend several results about degenerate scales previously found for the Laplace and Lamé equation to the biharmonic equation in 2 dimensions. The degenerate scales are obtained from a \(4\times 4\) discriminant matrix whose shape is provided for different kinds of domain symmetry. We show that degenerate scales can be obtained from a minimization problem. Then, we compare the degenerate scales of two boundaries, one included within the other. For smooth star-shaped curves, we show that there are only two degenerate scales, give sufficient conditions for not being at a degenerate scale and produce bounds to the degenerate scales. For symmetric smooth simply connected curves there are also only two degenerate scales. For these symmetric cases, the use of complex variables allows us to go further and to link the problem of biharmonic equation to the one of plane elasticity and to give information, including bounds and exact values of degenerate scales, for many cases, while until now only very few ones were known. Our results have some consequences for the biharmonic outer radius defined by Pólya and Szegö. The numerical computation of degenerate scales by using boundary elements confirms the theoretical results.

Spatial identification and multilevel zoning of land use functions improve sustainable regional management: a case study of the Yangtze River Economic Belt, China.

The quantitative identification and zoning management of land use functions (LUFs) are important starting points for solving the problems of resource allocation and sustainable development. In this study, with the Yangtze River Economic Belt (YREB) as the case study area, LUFs were grouped into three primary categories: economic function (ENF), social function (SCF), and ecological function (ELF). The least square error model was adopted to identify the morphological changes of LUFs. A two-dimensional discriminant matrix of the dynamic degree of LUF change and terrain niche index was constructed to explain the terrain gradient effect of LUFs. Bivariate local spatial autocorrelation was used to analyze the trade-offs in 2018 between ELF and ENF, and ELF and SCF. Finally, a new multilevel zoning scheme for LUFs was proposed. The results showed that from 1990 to 2018, ENF increased rapidly in cities along the Yangtze River, the overall level of SCF declined, and ELF in the south of the Yangtze River was better than that in the north. LUFs' morphological zoning exhibited significant regional differences. SCF-ELF combination areas and ELF dominance areas were mainly optimized in the second-level zoning. The areas with weak ELF were concentrated in the east of the YREB. Based on these results, nine kinds of LUF zonings and six kinds of major functional zonings were devised, and policy allocation was arranged for each zoning to improve the efficiency of spatial zoning management. Our research provides a reference for large-scale regional sustainable development and land use zoning management.

Initial-State Observability of Mealy-Based Finite-State Machine With Nondeterministic Output Functions

In mobile systems or the failure detection applications, the output for some input event is state-dependent and nondeterministic after intermittent sensor failures or measurement uncertainties, which does not hold under the conventional observability hypothesis. In this article, such cases can be modeled by a Mealy-based finite-state machine (FSM) with nondeterministic output functions, and we investigate the “initial-state” observability by use of matrix semitensor product (matrix-STP). First, to characterize the nondeterministic output functions, a virtual state set consisting of state–event pairs is introduced to obtain an augmented FSM. By resorting to the matrix-STP, the algebraic expression of augmented FSM is proposed. Subsequently, based on the newly constructed model, the initial-state observability can be verified by checking the distinguishability of state trajectories of the augmented FSM. Meanwhile, the necessary and sufficient condition for such initial-state observability is derived from a discriminant matrix consisting of polynomial elements. Finally, numerical examples show the validity of the proposed method. The current results are further conducive to explore the critical safety of cyber–physical systems in many real-world systems.

Locally generalized preserving projection and flexible grey wolf optimizer-based ELM for fault diagnosis of rolling bearing

It is a challenge for bearing fault diagnosis to effectively reduce the dimension of high dimensional data and improve the accuracy of fault identification. To address this issue, a novel fault diagnosis method based on locally generalized preserving projection (LGPP) and flexible grey wolf optimizer-based extreme learning machine (FGWO-ELM) is proposed in this paper. Firstly, the high dimensional features are obtained from the time–frequency domain of the signal. Secondly, a novel LGPP algorithm is proposed to reduce the dimension of high dimensional features more effectively. The algorithm uses a generalized discriminant matrix to calculate the degree similarity between data, rather than a simple Euclidean distance, so as to obtain sensitive features with better discrimination. Thirdly, a FGWO algorithm is proposed and used to optimize the parameters of ELM to improve the fault recognition rate. Finally, the effectiveness of the proposed fault diagnosis method is verified by two bearing experiments. Experimental results show that compared with other methods, the proposed method not only has the better dimension reduction ability, but also has higher diagnostic accuracy.

Sustainable Renewal of Historical Urban Areas: A Demand–Potential–Constraint Model for Identifying the Renewal Type of Residential Buildings

The assessment of buildings facilitates the exploration of the viability of progressive and organic renewal, prevents the aimless and unorganized demolition of residential buildings in previous urban renewal projects, and facilitates the balancing of the preservation and sustainable development of historical urban areas. Studies have rarely examined the effect of regional factors on building renewal or differentiated the effects of historic preservation and development conditions. The fundamental function of historical urban areas in China is to provide residence. Therefore, this study developed a framework to identify the renewal type of residential buildings. The developed framework was used to construct a demand–potential–constraint model with assessment indicators related to three dimensions, namely renewal demand, development potential, and preservation constraint. Moreover, discriminant matrices were employed to divide the renewal of residential buildings into four modes and subdivide it into six types. The historical urban area of Suzhou was selected as the study site. According to the results of renewal type identification, renewal schedules and models with high referential value can be developed by urban renewal planners, which facilitates the optimization of resource allocation. The developed framework provides novel theoretical and practical insights regarding building renewal assessment in historical urban areas.

Unsupervised Learning Discriminative MIG Detectors in Nonhomogeneous Clutter

Principal component analysis (PCA) is a commonly used pattern analysis method that maps high-dimensional data into a lower-dimensional space maximizing the data variance, that results in the promotion of separability of data. Inspired by the principle of PCA, a novel type of learning discriminative matrix information geometry (MIG) detectors in the unsupervised scenario are developed, and applied to signal detection in nonhomogeneous environments. Hermitian positive-definite (HPD) matrices can be used to model the sample data, while the clutter covariance matrix is estimated by the geometric mean of a set of secondary HPD matrices. We define a projection that maps the HPD matrices in a high-dimensional manifold to a low-dimensional and more discriminative one to increase the degree of separation of HPD matrices by maximizing the data variance. Learning a mapping can be formulated as a two-step mini-max optimization problem in Riemannian manifolds, which can be solved by the Riemannian gradient descent algorithm. Three discriminative MIG detectors are illustrated with respect to different geometric measures, i.e., the Log-Euclidean metric, the Jensen–Bregman LogDet divergence and the symmetrized Kullback–Leibler divergence. Simulation results show that performance improvements of the novel MIG detectors can be achieved compared with the conventional detectors and their state-of-the-art counterparts within nonhomogeneous environments.

城市低效用地更新为绿色基础设施优先度评价

PDF HTML阅读 XML下载 导出引用 引用提醒 城市低效用地更新为绿色基础设施优先度评价 DOI: 10.5846/stxb202108182283 作者: 作者单位: 作者简介: 通讯作者: 中图分类号: 基金项目: 国家自然科学基金项目（51478217） Priority evaluation of urban inefficient land renewal to green infrastructure Author: Affiliation: Fund Project: The National Natural Science Foundation of China (General Program) 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:在以存量更新促进城市可持续发展的时代，将城市低效用地有序更新为绿色基础设施（Green Infrastructure，GI）是实现城市建成区生态空间再造和生态格局优化的可操作路径。现有低效用地的更新优先度评价多将经济效益作为发展潜力的判断依据，未能很好地考虑地块的潜在生态效益。以广西桂林市中心城区为例，采用层次分析法、熵权法和TOPSIS方法，从经济和社会效益角度构建综合指标体系定量评价城市低效用地的更新潜力，在此基础上，基于最小费用路径和电路理论方法，采用标准化最小费用廊道值与中心性值，评价低效用地对提升研究区景观连通性的重要性，最后利用二维判别矩阵方法，综合更新潜力与重要性分析结果，对低效用地更新为GI的优先度进行了评判。结果表明：（1）研究区更新潜力等级较高的低效用地面积占比达38%，集中分布在研究区西部和北部。（2）低效用地更新后充当廊道和踏脚石的潜力存在显著差异，对提升景观连通性重要性较高的低效用地面积占比达40%，主要分布在研究区北部、西部和南部。（3）低效用地更新潜力评价结果和基于景观连通性提升的重要性评价结果在部分区域差异明显，更新为GI优先度较高的低效用地主要分布在研究区北部和西南部，应根据其不同优先度等级采取差异化的更新规划措施。研究结果可为城市低效用地再开发的规划实践和绿色基础设施网络格局优化提供新框架和决策参考。 Abstract:In the period of promoting urban sustainable development by stock renewal, orderly renewal of urban inefficient land to green infrastructure (GI) is an operational path to realize the reconstruction of ecological space and the optimization of ecological pattern in urban built-up area. In the existing researches, the evaluation of renewal priority of urban inefficient land mostly takes economic benefits as a basis of judging the development potential, and fails to consider the potential ecological benefits of the land parcels. Taking the downtown area of Guilin city as an example, this paper constructed a comprehensive index system to quantitatively evaluate the renewal potential of urban inefficient land from the perspective of economic and social benefits by using analytic hierarchy process, entropy weight method and TOPSIS method. Then, using the Least-Cost Path and the Circuit Theory method, we adopted the normalized least cost corridor value and the centrality value to evaluate the relative importance of urban inefficient land for the improvement of landscape connectivity in the study area. Finally, using two-dimensional discriminant matrix method, the priority of urban inefficient land renewal to GI was evaluated by integrating the results of renewal potential evaluation and importance analysis. The results showed that:(1) the area of inefficient land parcels with high renewal potential accounted for 38%, which were relatively concentrated in the west and north of the study area. (2) The renewal of inefficient land to GI could significantly improve the overall landscape connectivity of the study area, and for these inefficient land parcels, there were significant differences in their potential of serving as corridors and stepping stones after renewal. The area of inefficient land parcels which were highly important for improving landscape connectivity accounted for 40%, and they were mainly distributed in the north, west and south of the study area. (3) There were obvious differences between the results of inefficient land renewal potential evaluation and the results of importance analysis based on the improvement of landscape connectivity in some regions. The area of inefficient land parcels with high priority for renewal accounted for 21%, and they were mainly distributed in the north and southwest of the study area. For these inefficient land parcels, different renewal planning measures should be taken according to their different renewal priority grades. The research results can provide a new method framework and decision-making reference for the planning practice of urban inefficient land redevelopment and the optimization of green infrastructure network patterns. 参考文献 相似文献 引证文献

Visual analytic hierarchical process for in situ identification of leakage risk in urban water distribution network

Assessment and early warning of leakage risk in the water distribution network are important for securing urban sustainable development. However, the identification of high leakage areas in urban water distribution networks lacks visual model-driven approaches. In the current study, the model parameters of the water distribution network (i.e., leakage point density, pipe age and material, road load, and average pressure of the pipe network) were collected in Haidian Island, Hainan Province, China as a case. The virtual district metering area (V-DMA) was used to modularize the water distribution network. A visual analytic hierarchical process (V-AHP) based on the V-DMA was established for identifying the leakage high-risk areas in the V-DMA. The first and second-level discriminant matrices (i.e., subjective and objective weights, respectively) were calculated based on the model parameters. Integration of V-AHP with V-DMA reveals insight into high leakage areas of urban water distribution networks, which was consistent with in-situ observations. Overall, this study provides a simple, efficient, and practically applicable method for identifying the leakage areas in water distribution networks.

Cultivated Land Use Zoning Based on Soil Function Evaluation from the Perspective of Black Soil Protection

Given that cultivated land serves as a strategic resource to ensure national food security, blind emphasis on improvement of food production capacity can lead to soil overutilization and impair other soil functions. Therefore, we took Heilongjiang province as an example to conduct a multi-functional evaluation of soil at the provincial scale. A combination of soil, climate, topography, land use, and remote sensing data were used to evaluate the functions of primary productivity, provision and cycling of nutrients, provision of functional and intrinsic biodiversity, water purification and regulation, and carbon sequestration and regulation of cultivated land in 2018. We designed a soil function discriminant matrix, constructed the supply-demand ratio, and evaluated the current status of supply and demand of soil functions. Soil functions demonstrated a distribution pattern of high grade in the northeast and low grade in the southwest, mostly in second-level areas. The actual supply of primary productivity functions in 71.32% of the region cannot meet the current needs of the population. The dominant function of soil in 34.89% of the area is water purification and regulation, and most of the cultivated land belongs to the functional balance region. The results presented herein provide a theoretical basis for optimization of land patterns and improvement of cultivated land use management on a large scale, and is of great significance to the sustainable use of black soil resources and improvement of comprehensive benefits.

Enhanced Palmprint Identification Using Score Level Fusion

Palm print identification has been used in various applications in several years. Various methods have been proposed for providing biometric security through palm print authentication. One such a method was feature level fusion which used multiple feature extraction and gives higher accuracy. But it needed to design a new matcher and acquired many training samples. However, it cannot adapt to scenarios like multimodal biometric, regional fusion, contactless and complete direction representation. This problem will be overcome by score level fusion method. In this article, we propose a salient and discriminative descriptor learning method (SDDLM) and gray-level co-occurrence matrix (GLCM).The score values of SDDLM and GLCM are integrated using score level fusion to provide enhanced score. Experiments were conducted on IITD palm print V1 database. The combination of SDDLM AND GLCM methods will be useful in achieving higher performance. It provides good recognition rate and reduces computation burden.

Evaluation algorithm of alhagi sparsifolia desertification control under different irrigation amounts

Desertification control is an important issue that must be considered in modern society. In order to effectively improve the accuracy and practicability of the evaluation algorithm of desertification control effect, the Alhagi sparsifolia index under different irrigation amount was taken as the research object, and the evaluation algorithm of desertification control effect was proposed. In the “vegetation-sandstorm-soil” index system, a number of indexes were selected according to the core environmental parameters of Alhagi sparsifolia and grassland desertification. And the analytic hierarchy process, remote sensing, geographic information system, and landscape technology were used to assign index weights of desertification control capacity, which were calculated by multiple discriminant matrices. Finally, the data regression analysis was performed based on remote sensing and computer image information screening and processing to determine the final evaluation results. The experimental data show that the true positive rate of the algorithm in this paper is between 160 and 200, which is within a large range of advantages, indicating that the overall evaluation accuracy of the algorithm is high and the evaluation effect is perfect.

Sparse Graph Embedding Based on the Fuzzy Set for Image Classification

In recent years, many face image feature extraction and dimensional reduction algorithms have been proposed for linear and nonlinear data, such as local-based graph embedding algorithms or fuzzy set algorithms. However, the aforementioned algorithms are not very effective for face images because they are always affected by overlaps (outliers) and sparsity points in the database. To solve the problems, a new and effective dimensional reduction method for face recognition is proposed—sparse graph embedding with the fuzzy set for image classification. The aim of this algorithm is to construct two new fuzzy Laplacian scattering matrices by using the local graph embedding and fuzzy k-nearest neighbor. Finally, the optimal discriminative sparse projection matrix is obtained by adding elastic network regression. Experimental results and analysis indicate that the proposed algorithm is more effective than other algorithms in the UCI wine dataset and ORL, Yale, and AR standard face databases.

Retinal Vessel Segmentation based on Convolutional Neural Network and Connection Domain Detection

We propose a retinal vessel segmentation algorithm based on a convolution neural network and connected domain detection. After defining the discriminant matrix, we constructed and trained a convolutional neural network model, which can realize the mapping relationship from eye fundus grayscale to the discriminant matrix. This model achieves the preliminary segmentation of retinal vessels. The prediction of uncertain pixels is revised by using the geometric characteristics of the vessels and through the analysis of connected regions. The experimental results show good generalization ability, the average segmentation accuracy, specificity, and sensitivity are 96.64%, 97.96%, and 80.68%, respectively.